We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Unbounded Gradients in Federated Learning with Buffered Asynchronous Aggregation

Abstract: Synchronous updates may compromise the efficiency of cross-device federated learning once the number of active clients increases. The \textit{FedBuff} algorithm (Nguyen et al., 2022) alleviates this problem by allowing asynchronous updates (staleness), which enhances the scalability of training while preserving privacy via secure aggregation. We revisit the \textit{FedBuff} algorithm for asynchronous federated learning and extend the existing analysis by removing the boundedness assumptions from the gradient norm. This paper presents a theoretical analysis of the convergence rate of this algorithm when heterogeneity in data, batch size, and delay are considered.
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Optimization and Control (math.OC); Machine Learning (stat.ML)
Cite as: arXiv:2210.01161 [cs.LG]
  (or arXiv:2210.01161v1 [cs.LG] for this version)

Submission history

From: Mohammad Taha Toghani [view email]
[v1] Mon, 3 Oct 2022 18:20:48 GMT (674kb,D)

Link back to: arXiv, form interface, contact.